VIEWS ON NEUTRONICS AND ACTIVATION ISSUES FACING LIQUID-PROTECTED IFE CHAMBERS
L. El-Guebaly and the ARIES Team
University of Wisconsin, Fusion Technology Institute, 1500 Engineering Dr., Madison, WI
elguebaly@engr.wisc.edu
0.5 m
Target
| Nozzles |
Shield
During the early 2000s, the ARIES team defined the
design space and operational windows for both laser and
heavy ion driven inertial IFE concepts from the viewpoint
of a viable power plant system, rather than developing a
point design1. One of the ARIES tasks addressed the
nuclear-related issues for thick-liquid protected chambers
with heavy ion (HI) drivers. The outcome of the nuclear
study2 is the subject of this paper. Previous publications
have assessed the radiological issues for the thin liquid
wall concept3 and the HI target debris4.
3 m Radius
|
Gap
I. INTRODUCTION
rewelding the structure and flow nozzles during operation
for component replacement or maintenance. Moreover,
the activation of the steel-based chamber structure could
be severe. Another concern is related to the IFE pulsed
nature. The high instantaneous damage rate and
deposition of the neutron energy can change the
microstructure and cause isochoric heating problems. The
following sections cover in more detail all these issues
and highlight the main findings of our nuclear assessment
for the ARIES-IFE thick-liquid-protected chamber.
Liquid Jets
Thin and thick liquid walls provide an attractive
solution to the challenging material issues facing the
heavy-ion applications of the inertial fusion energy (IFE)
concept. Given the many advantages of liquid-protected
chambers, there are several nuclear-related concerns that
are discussed in detail for the thick liquid wall option in
particular. These are the ability to protect the steel-based
structure from radiation damage and high activation, the
feasibility of rewelding the structure, and the pulserelated problems. These issues have a profound impact on
the ARIES-IFE thick-liquid protected chamber design.
Chamber Wall
A representative radial build for the near-spherical
thick-liquid protected chamber is given in Fig. 1, showing
a thick-liquid wall surrounded by the shield. The
innermost layer of the shield represents the location of the
chamber wall and flow nozzles. Two candidate liquid
metal breeders have been considered for this study: Flibe
(BeF2,[LiF]2) and Flinabe (NaF, LiF, BeF2). The latter has
a substantially lower melting point5 compared to Flibe
(459 oC), offering lower operating temperature and vapor
pressure6.
Here, we are concerned with the feasibility of
protecting the steel-based structure of the chamber with a
thick liquid wall. Specifically, the concern is the ability of
the liquid wall to protect the structure during the entire
plant life while providing an adequate tritium supply for
machine operation and satisfying the ARIES top-level
requirement of generating only low-level waste. In
addition, the helium production level at the chamber
structure is a concern if the design mandates cutting and
Fig. 1. Schematic of ARIES-IFE-HIB radial build.
II. MODEL, DESIGN CRITERIA, AND LIMITS
The 460 MJ yield HIB targets are repetitively
injected into the chamber at a rep rate of 4 Hz. Following
target injection, multiple HI beams focus on the target,
emitting intense x-rays that compress the DT capsule and
initiate the fusion process, generating energetic neutrons
and ions. During burn, the 14.1 MeV neutrons experience
several collisions with the dense capsule materials and
lose a fraction of their energy. The average neutron
energy is 11.8 MeV. After burn, the x-rays, neutrons, ions
and target debris travel through the cavity, reaching the
liquid wall in less than a microsecond. The chamber
clearing system pumps out the liquid along with the target
debris and reconditions the chamber in time for the next
shot.
Our computational model included the liquid wall
and shield as arranged in Fig. 1 and explicitly considered
the projected 85% system availability. Approximately 126
million pulses per full power year (FPY) have been
modeled to estimate the activation level of the structure
using the ALARA pulsed activation code7. The neutron
flux throughout the chamber was calculated with the
DANTSYS8 discrete ordinates transport code and the
FENDL-2 175-neutron 42-gamma group coupled crosssection library9. Table I summarizes the design criteria
and limits. A tritium breeding ratio (TBR) of 1.08 assures
tritium self-sufficiency for ARIES-IFE-HIB. Higher
breeding should be avoided as the regulatory agency may
not license fusion devices with excess tritium production.
Besides breeding tritium and recovering the energy, the
liquid wall has an important shielding function. It
provides a lifetime protection for the shield. The lifelimiting criterion for steel-based shielding components
has traditionally been the displacement of atoms, ranging
between 25 dpa for austenitic steel to 200 dpa for
advanced ferritic steel (FS). Another design criterion
relates to the reweldability of the structure and flow
nozzles. There is always a need to cut and reweld these
components for maintenance and replacement reasons.
The helium production at the innermost surface of the
shield should remain below the reweldability limit of 1
appm at any time during plant operation.
As a top-level requirement for ARIES power plants,
all components should generate only low-level waste
(LLW) and meet both Fetter’s10 and 10CFR61 NRC11
waste disposal limits for Class C wastes. A computed
volumetric average waste disposal rating (WDR) < 1 at
the end of a 100-year institutional control period at the
disposal site means the component qualifies for shallow
land burial as LLW. We evaluated the WDR for both
Fetter’s and NRC limits and reported the highest value.
1.3
Overall TBR
1.2
1.1
1.0
Required
TBR
Flibe
Flinabe
0.9
0.8
50
100
150
200
Liquid Wall Thickness (cm)
Fig. 2. Sensitivity of TBR to liquid wall thickness.
TABLE I. Design Criteria and Limits
Criteria
Overall TBR
dpa to advanced FS structure
dpa to 304-SS structure
He production at structure
WDR
Limit
1.08
200 dpa
25 dpa
1 He appm
1
III. RESULTS
III.A. Tritium Breeding
We started the analysis by examining the breeding
capacity of the candidate breeders using natural lithium,
then defined the breeder parameters in terms of thickness
and Li enrichment. The homogenized liquid wall consists
of 58% breeder and 42% void. It is found that Flibe is a
better breeder than Flinabe, thus a wall thickness of 85 cm
is required for Flibe and 150 cm for Flinabe for a
breeding ratio of 1.08 (refer to Fig. 2). The reported
overall TBR takes into account the 3% FW coverage for
penetrations and the tritium bred in the shield. The Li
enrichment did not enhance the breeding potential of the
relatively thick Flinabe wall. Most of the neutron power is
deposited in the liquid wall. The heat leakage to the shield
is 10% for the Flibe system and 1% for the Flinabe
system. The overall neutron energy multiplication
amounts to ~1.25 for both systems.
III.B. Radiation Damage to Shield
The ARIES team has assessed two structural
materials for the thick liquid protected chamber:
austenitic steel and low-activation FS12. Over the past 10
years, 304-SS has been the material of choice for the
HYLIFE design13 for being highly corrosion-resistant to
Flibe. The new class of low-activation FS alloys (such as
ODS-MF82H14) offer higher tolerance for radiation
damage, lower neutron-induced swelling, lower thermal
expansion coefficient, and higher range of operating
temperatures.
The shield protects the externals, operates hot to
recover the leaked energy from the liquid wall, and
contains 10% Flibe/Flinabe coolant. Figure 3 illustrates
the drop of the damage at the shield with the liquid wall
thickness. The 200 dpa level at the shield indicates both
breeders (85 cm thick Flibe jets and 150 cm thick Flinabe
jets) could protect the advanced FS structure for 40 FPY.
Switching from FS to 304-SS structure would require a
substantial increase in the Flibe jet thickness (from 85 cm
to 130 cm). If so, the system breeds excess tritium that
raises safety and licensing concerns. A potential solution
would be to deplete the lithium of the Flibe at an extra
cost.
III.D. Activation and WDR of Shield
4
Peak dpa to Structure
(dpa @ 40 FPY)
10
3
10
FS Limit
2
10
Flibe
Flinabe
304-SS Limit
1
10
0
10
50
100
150
200
Liquid Wall Thickness (cm)
Fig. 3. Variation of radiation damage at the shield inner
surface with liquid wall thickness.
III.C. Helium Production Level at Shield
Even though the 200 dpa damage limit can be met,
the reweldability limit (1 He appm) is greatly exceeded,
meaning the chamber structure and flow nozzles cannot
be rewelded if there is a need to cut and reweld the
structure at any time during operation. Figure 4 illustrates
the severity of the problem. The alternate option of
replacing the shield and nozzles with new components in
case of a failure must be carefully examined as
replacement could be a prohibitive expense and require
extended shutdown of the machine. The welds should be
hidden and located away from the high-radiation zones.
Peak Helium Production at
Structure (appm @ 40 FPY)
4
10
3
10
2
10
Flibe
Flinabe
1
10
0
10
Reweldability Limit
-1
10
50
100
150
200
Liquid Wall Thickness (cm)
Fig. 4. Sensitivity of He production at front layer of shield
to thickness of liquid wall.
We examined the two candidate steel alloys:
advanced low-activation M-F82H FS and austenitic 304SS. The alloying elements and main impurities from
References 14 and 15 are given in Table II. Both alloys
generate high-level waste, particularly the inner shield
surface and flow nozzles. 304-SS generates very high
level waste as Figures 5 and 6 indicate for both Flibe and
Flinabe systems. The HLW violates the ARIES top-level
waste requirement that calls for only LLW (WDR < 1).
The main contributors to the WDR are 94Nb (from Nb),
99
Tc (from Mo), and to a lesser extent, 192nIr (from W).
One would expect to solve the FS waste problem by
averaging the WDR over a thicker shield or controlling
the Nb and Mo impurities. The ODS-MF82H-FS (the
leading candidate for the ARIES-IFE-HIB chamber) can
meet the Class C limit with at least 50 cm thick shield. Nb
and Mo impurity control is a “must” requirement for the
ODS-MF82H-FS structure of the Flibe system. Even
though tailoring the FS materials to reduce the Nb and Mo
concentrations helps reduce the WDR, practically and
economically, complete removal of these materials can
never be accomplished, requiring shield thicker than 50
cm. Two solutions have been identified to solve the
nozzle waste problem: thickening the blanket, or
preferably, mixing the nozzles with the shield (if
acceptable by the regulatory agency) and disposing them
as a single unit after plant decommissioning. The former
solution could be prohibitively costly as it calls for a
larger chamber with depleted Flibe/Flinabe.
TABLE II. Composition of Steel Alloys (in wt%)
Elements
Fe
C
N
O
Si
P
S
Ti
V
Cr
Mn
Co
Ni
Cu
Nb
Mo
Ta
W
Y
ODS M-F82H-FS*
87.981
0.04
0.005
0.13
0.24
0.005
0.002
0.09
0.29
8.7
0.45
0.0028
0.0474
0.01
0.00033
0.0021
0.08
2
0.7
304-SS
70.578
0.046
0.038
-0.47
0.026
0.012
0.03
-17.7
1.17
0.1
9.3
0.2
-0.33
----
* M-F82H-FS + 0.25wt% Y2O3. Other impurities include B,
Al, As, Pd, Ag, Cd, Sn, Sb, Os, Ir, Bi, Eu, Tb, Dy, Ho, Er, U.
 The heating diminishes in microseconds.
3
Nozzles
2
|
10
2
304-SS
10
1
10
ODS-MF82H-FS
0
10
Class C Limit
ODS-MF82H-FS
without Nb and Mo
-1
10
0
20
40
60
80
Shield Thickness (cm)
100
Fig. 5. WDR of structure as a function of shield thickness
behind 85 cm thick Flibe liquid wall.
III.E. Isochoric Heating
WDR of Shield Structure
WDR of Shield Structure
10
|
1
10
304-SS
Nozzles
ODS-MF82H-FS
Class C Limit
0
10
-1
10
ODS-MF82H-FS
without Nb and Mo
-2
10
0
20
40
60
80
Shield Thickness (cm)
100
Fig. 6. WDR of structure as a function of shield thickness
behind 150 cm thick Flinabe liquid wall.
To understand the isochoric heating problem, it is
essential to identify the evolution of the liquid wall with
time following the target implosion. When the x-rays
rapidly deposit their energy in the liquid, they vaporize a
few microns and drive strong shock waves into the liquid.
The neutrons deposit their energy volumetrically in the
liquid and the underlying structure. The geometry of the
liquid hardly changes before the arrival of the neutrons.
The neutron heating causes pressurization and rapid
expansion of the liquid. The hydromotion leads to
splashing of the liquid and breakup of the entire liquid
wall or jets. The disassembly of the liquid wall/jets is
allowed only in free-jet systems, like the HYLIFE-II thick
liquid wall design13. In thin-liquid wall designs16, there is
no splashing as the liquid is contained in a porous
structure and the pressure is lower because the first wall
radius is placed at much larger radius than 0.5 m.
The survivability of the structure over millions of
shots is still an open question. The instantaneous
deposition of the neutron energy can cause isochoric
heating problems with significant pressure waves that
could impact the fatigue life of the structure. The structure
will be heated and cooled at the same rate as the target rep
rate (4 Hz). This means the structure temperature
fluctuates four times per second with internal pressure
reaching 100 atm. The sudden deposition of the nuclear
heating and the fluctuation of the structure temperature
can induce high stress/strain in the FS structure. Fatigue
from cycling and repetitive shock waves could shorten the
structure lifetime and cause internal cracks. When
combined with the neutron-induced radiation damage
(such as displacement of atoms), the fatigue life of the
steel becomes a concern particularly for designs with high
nuclear heat loads.
The fusion reactions occur during a very short burn
time (10-100 ps). Most of the high-energy neutrons reach
the liquid surface in 10-150 ns, depending on the surface
radius. The lower energy neutrons arrive over a longer
period of time. The neutrons spend tens of nanoseconds
slowing down within the liquid blanket. Over the past 30
years, only the HIBALL study16 has performed rigorous
time-dependent nuclear heating analysis for the liquid and
structure17. By inspecting the results, several broad trends
can be described:
 The heating has a sharp peak, then quickly decays
IV. CONCLUDING REMARKS
 The peak is several orders of magnitude above the timeaveraged power density
 The temporal distribution has a narrow width of ~20 ns
No breeding problem has been found for the ARIESIFE thick-liquid protected concept. However, we
identified several serious nuclear problems that need
further consideration. These relate to waste level,
structural integrity of the chamber, and economics.
The activation of the chamber structure is severe. The
proposed design generates high-level wastes that violate
the ARIES top-level waste requirement. Note that this
waste problem is generic to the thick-liquid concept as a
similar problem has been identified for the HYLIFE
design in the mid-1990s18. The apparent best solution to
the waste problem is a combination of protecting the
shield with thicker Flibe (or Flinabe) liquid walls with
depleted lithium, adjusting the size of the shield itself, and
more importantly, controlling the Nb and Mo impurities
of the FS structure. Thus, consideration of advanced FS
for ARIES-IFE-HIB rests heavily on the assumption that
Nb and Mo impurities can be drastically controlled for the
M-F82H structural material. The high cost of impurity
control must be factored in the unit cost of the modified
FS structure. Moreover, the cost-of-electricity should
reflect the economic penalty associated with the larger
chamber and depleted Flibe/Flinabe breeders. Besides the
economic concerns, the formation and stability of the
sizable liquid jets (> 85 cm thick) are feasibility issues
under debate in the fusion community.
With regard to the radiation damage issue, while the
liquid jets control the FS atomic displacements below the
200 dpa limit needed for structural integrity, the neutroninduced helium production level is excessive and
precludes the reweldability of the chamber FS stucture
and flow nozzles during plant operation. No viable
solution has been found for the reweldability problem.
We hope, however, that the materials community can
develop a more forgiving FS with three orders of
magnitude higher reweldibility limit to avoid the
economic penalty of shutting down the machine for
extended periods to replace the chamber structure and
flow nozzles. In relation to the pulsed nature of the
concept, the high instantaneous damage rate can lead to
significant changes in the microstructure of the structure.
The repetitive shock waves, the instantaneous deposition
of the neutron energy, and the fluctuation of structure
temperatures will certainly impact the fatigue life and
produce significant stress/strain and internal cracks. This
could shorten structure lifetime making the structure
fatigue an important life-limiting factor as the 200 dpa
limit. In future thick liquid wall studies, it is essential to
address the combined effect of radiation damage and
fatigue on the lifetime of the chamber structure.
ACKNOWLEDGMENTS
This work was performed under the auspices of the
US DOE (contract # DE-FG02-98ER 54462).
[1]
[2]
[3]
REFERENCES
F. NAJMABADI, R. RAFFRAY et al.,
“Operational Windows for Dry-Wall and
Wetted-Wall IFE Chambers,” Fusion Science &
Technology, 46, No. 3, 401 (2004).
L. EL-GUEBALY, “Ferritic Steel Assessment
and Self-Consistent Nuclear Parameters for
ARIES-IFE-HIB,” Presentation given at ARIES
Project Meeting (October 2002). Available at:
http://aries.ucsd.edu/ARIES/.
L.
EL-GUEBALY,
P.
WILSON,
D.
HENDERSON, L. WAGANER, and R.
RAFFRAY, “Radiological Issues for Thin
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
Liquid Walls of ARIES-IFE Study”, Fusion
Science & Technology, 44, 405 (2003).
L.
EL-GUEBALY,
P.
WILSON,
D.
HENDERSON, and A. VARUTTAMASENI,
“Feasibility of Target Materials Recycling as
Waste Management Alternative”, Fusion Science
& Technology, 46, No. 3, 506 (2004).
P. PETERSON, University of CaliforniaBerkeley, Private Communications (September
2002).
T. LUTZ, “Measurement of the Melting Point
Temperature of Several Lithium-SodiumBeryllium Fluoride Salt (FLiNaBe) Mixtures,”
These proceedings.
P. WILSON and D. HENDERSON, “ALARA:
Analytic and Laplacian Adaptive Radioactivity
Analysis Code Technical Manual,” University of
Wisconsin
Fusion
Technology
Institute,
UWFDM-1070 (January 1998).
DANTSYS: A Diffusion Accelerated Neutral
Particle Transport Code System, Los Alamos
National Laboratory Report, LA-12969-M
(1995).
Available at: http://www.iaea.org/.
S. FETTER, E. T. CHENG, and F. M. MANN,
“Long Term Radioactive Waste from Fusion
Reactors: Part II,” Fusion Engineering and
Design, 13, 239 (1990).
Nuclear Regulatory Commission, 10CFR61,
Licensing Requirements for Land Disposal of
Radioactive Waste, Federal Register, FR47,
57446 (1982).
M. BILLONE, R. RAFFRAY, D.K. SZE, and L.
EL-GUEBALY, “ARIES Assessment of IFE
Structural Material,” University of California
San Diego Report UCSD-ENG-101 (2002).
R. MOIR, R. BIERI, X. CHEN et al., “HYLIFEII: a Molten-salt Inertial Fusion Energy Power
Plant Design - Final Report,” Fusion
Technology, 25, 5 (1994).
M. BILLONE, Argonne National Laboratory,
Private Communications (September 2002).
C. BAKER et al., “Starfire-A Commercial
Tokamak Fusion Power Plant Study,” Argonne
National Laboratory Report, ANL/FPP-80-1
(1980).
B. BADGER, F. ARENDT, K. BECKER et al.,
“HIBALL - A Conceptual Heavy Ion Beam
Driven Fusion Reactor Study,” University of
Wisconsin Fusion Technology Institute Report,
UWFDM-450 (September 1981).
M. SAWAN, G. MOSES, and G. KULCINSKI,
“Time Dependent Neutronics Analysis for the
HIBALL Heavy Ion Beam Fusion Reactor,”
Nuclear Technology/Fusion, 2, 215 (1982).
[18]
J.D. LEE, “Waste Disposal Assessment of
HYLIFE-II Structure,” Fusion Technology, 26,
74 (1994).